Author Topic: MOSFET internal diode  (Read 18176 times)

0 Members and 1 Guest are viewing this topic.

Offline allanwTopic starter

  • Frequent Contributor
  • **
  • Posts: 343
    • Electronoblog
MOSFET internal diode
« on: June 25, 2010, 11:52:36 pm »
I'm using one of these as a switch for a small electromagnet (200mA at 5V):

http://www.irf.com/product-info/datasheets/data/irl530n.pdf

I believe electromagnets are slightly inductive loads, so I'll need a diode to prevent back-emf damage. However, it seems like the MOSFET I have already has an integrated diode going from source to drain. Does that mean I can just wire the drain to ground, source to the load (which is flipped from how I'd normally wire a MOSFET) and have this integrated diode do the same job?

Example circuits online show their diodes are in parallel, not in series, as this setup would be. Would it still do the job?


« Last Edit: June 25, 2010, 11:58:53 pm by allanw »
 

Offline Pyr0Beast

  • Contributor
  • Posts: 47
Re: MOSFET internal diode
« Reply #1 on: June 26, 2010, 01:02:26 am »
Don't use the mosfet's parasitic diode as a real one since it is slow as hell.

If you need to do that, you can get mosfets with internal diode.

Or use an external one.

It would be good to omit zener diode, these things can cause ringing on the gates. If you are worried about excess voltage on gate, use it at the driver part.


Electromagnets are inductive loads, same with relays.

Clamping diode is installed (paralel) at the inductive load. A simple diode in series doesn't do a thing.

EMF can be reduced with ferrite rings, chokes and snubbers.

You can install a ferrite ring on the drain of a mosfet and a good snubber across it - 100nF multilayer cap with ~20-100 Ohm resistor in series

 

Offline allanwTopic starter

  • Frequent Contributor
  • **
  • Posts: 343
    • Electronoblog
Re: MOSFET internal diode
« Reply #2 on: June 26, 2010, 01:34:54 am »
Hmm, I think I measured the inductance of my electromagnet at 50mH to 100mH at 100Hz. That seems like a lot. Simulating this in LTspice, there's a 100V spike. I'll put a diode in parallel then.
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18085
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: MOSFET internal diode
« Reply #3 on: June 26, 2010, 06:50:59 am »
I think the simulation may produce larger spikes as its working on theory and not practical parts, however yes you need to be careeful, as other said don't use the mosfets diode for this i tried it once and it made the thing run hotter and it is not designed for it, it is just for emergency only, I was doing some testing on an air con system at work recently where the "engineer" did not realize he needed diodes on ALL inductive loads (some had diodes) and i got up to 400 V out of the thing, infact i got a shock off it.
 

Offline Pyr0Beast

  • Contributor
  • Posts: 47
Re: MOSFET internal diode
« Reply #4 on: June 26, 2010, 10:22:51 am »
Any inductance will degrade mosfet over time if left untreated, same with transistors, ic's and other silicone chips.
Inductive kickback can go all the way up and downstream, causing all sorts of weird errors, like pushing powersupply into overvoltage protection, perhaps causing it to oscillate wildly, reset or freeze uC's, etc.

IRL530n is a good part at such a small load. Perhaps watch gate capacitance if you see gate going up and down not fast enough, such as the mosfet is in its linear region, heating wildly.



Even some old bells, working at about 6V DC could give you a nasty shock.
 

Offline allanwTopic starter

  • Frequent Contributor
  • **
  • Posts: 343
    • Electronoblog
Re: MOSFET internal diode
« Reply #5 on: June 26, 2010, 02:06:32 pm »
Yeah I decided to use the IRL530N because they're ridiculously cheap at digikey right now, due to them phasing out the non-RoHS version. It's only $0.63 in single quantities, while the replacements are $1.42. I stocked up on enough to last me a while :P
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20109
  • Country: gb
  • 0999
Re: MOSFET internal diode
« Reply #6 on: June 27, 2010, 05:31:02 pm »
The IRL540 is better because it has a lower channel resistance.
 

Offline scrat

  • Frequent Contributor
  • **
  • Posts: 608
  • Country: it
Re: MOSFET internal diode
« Reply #7 on: June 30, 2010, 04:45:17 pm »
Which is the circuit you are simulating? Is it similar to the one you posted at first? If it is so, the intrinsic diode has no influence , since it direction is opposite to the current on the "motor" when this last is energized
One machine can do the work of fifty ordinary men. No machine can do the work of one extraordinary man. - Elbert Hubbard
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18085
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: MOSFET internal diode
« Reply #8 on: June 30, 2010, 05:16:29 pm »
yes but when you turn that motor off you will have a back EMF spike that will be of reverse polarity and if not for that diode will destroy the mosfet
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20109
  • Country: gb
  • 0999
Re: MOSFET internal diode
« Reply #9 on: June 30, 2010, 10:33:28 pm »
Yes, the parasitic diode is in the wrong place to be effective for back EMF suppression so another diode is required.
 

Offline Time

  • Frequent Contributor
  • **
  • Posts: 725
  • Country: us
Re: MOSFET internal diode
« Reply #10 on: June 30, 2010, 10:48:07 pm »
Its called a flyback or free wheeling diode.  Its a pretty good rule of thumb to have one when switching any considerable amount of inductance.
-Time
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18085
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: MOSFET internal diode
« Reply #11 on: July 01, 2010, 05:30:08 am »
Its called a flyback or free wheeling diode.  Its a pretty good rule of thumb to have one when switching any considerable amount of inductance.

Thats what I tried to explain to them at work when I found 400 V back EMF in their air conditioning system,
 

Offline Pyr0Beast

  • Contributor
  • Posts: 47
Re: MOSFET internal diode
« Reply #12 on: July 01, 2010, 08:35:50 pm »
Its called a flyback or free wheeling diode.  Its a pretty good rule of thumb to have one when switching any considerable amount of inductance.

Thats what I tried to explain to them at work when I found 400 V back EMF in their air conditioning system,

Don't worry. People usually are like this.
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18085
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: MOSFET internal diode
« Reply #13 on: July 01, 2010, 09:36:45 pm »
yea particularly when its the company's flagship contract that they are putting down the toilet !, we had an item returned the other day that is a few years old, guess what ? it has relays with internal diodes, the twit doing the new stuff has only just discovered that these exist  ???
 

Offline Pyr0Beast

  • Contributor
  • Posts: 47
Re: MOSFET internal diode
« Reply #14 on: July 03, 2010, 09:45:14 pm »
yea particularly when its the company's flagship contract that they are putting down the toilet !, we had an item returned the other day that is a few years old, guess what ? it has relays with internal diodes, the twit doing the new stuff has only just discovered that these exist  ???

At least he learned, not many do :)
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18085
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: MOSFET internal diode
« Reply #15 on: July 03, 2010, 11:31:40 pm »
Ah yes but i had to tell him to put fly-back diodes on fan motors, he thought relays needed one but did not seem to realize fan motors do ..... duuuuuuuuuuuuuuuuuuuuuuuuh
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20109
  • Country: gb
  • 0999
Re: MOSFET internal diode
« Reply #16 on: July 04, 2010, 12:29:23 am »
I would;ve thought the motors would be AC on an air conditioning system so ordinary diodes wouldn't work, a snubber network, a MOV or a couple of zeners back-to-back would be the only options.
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 18085
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: MOSFET internal diode
« Reply #17 on: July 04, 2010, 06:50:04 am »
24 V vehicle supply
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf